Prior to amyloidosis (at 6 months of age), HCHF diet elicited a decrease in hippocampal glucose uptake and an attenuation of resting perfusion; with heightened functional hyperemia. In the established AD stage (12 months of age), HCHF-fed TgAD rats displayed increased hippocampal glucose uptake, and potentiated functional hyperemia in relation to CHOW nTg. With sustained diet exposure over the subsequent three months of disease progression (i.e., at 15 months of age), hippocampal glucose uptake of HCHF-fed TgAD rats was indistinguishable from that of obese 15-month old nTg rats.
Immunofluorescence and immunohistochemistry
Representative tomato lectin-Texas Red stained images of hippocampal capillaries in CHOW- and HCHF-fed nTg and TgAD 12-month-old rats are shown in Fig. 3.A. Hippocampal capillary diameter of 12-month-old CHOW-fed nTg rats was greater than that of CHOW-fed TgAD rats (by 0.79 ± 0.11 µm, p < 0.001), than HCHF-fed nTg rats (by 0.66 ± 0.12 µm, p = 0.002), and than HCHF-fed TgAD rats (by 0.63 ± 0.11 µm, p = 0.002) (Fig. 3.B) showing significant interaction of diet and genotype (p < 0.001) and diet effects (p < 0.001). Despite having no significant sex effect, capillary diameter of CHOW-fed female TgAD rats are significantly lower than HCHF-fed female nTg rats (by -0.94 ± 0.14 µm, p < 0.001). HCHF diet reduced the capillary diameter of both female (by -0.66 ± 0.14 µm, p = 0.019) and male (by -0.67 ± 0.17 µm, p = 0.047) nTg rats and in male TgAD rats compared to CHOW-fed nTg rats (by -0.72 ± 0.16 µm, p = 0.019) (Fig. 3.C).
Hippocampal lectin density (Fig. 4) of 12-month-old TgAD rats was greater compared to nTg rats (by 6.0% ± 0.65%, p < 0.001 in CHOW-fed cohorts and by 4.0% ± 0.61%, p < 0.001 in HCHF diet-fed cohorts) (Fig. 4.B.i). With prolonged HCHF diet at 15 months, lectin density of HCHF diet-fed TgAD rats significantly decreased (by -2.9% ± 0.57%, p < 0.001) at the level of HCHF diet-fed nTg rats. The differences in lectin density at 12 months between genotypes and the effect of prolonged HCHF diet in TgAD rats was mainly evident when only female rats were considered (p = 0.004) (Fig. 4.B.ii).
In addition, from the stained images of hippocampal AQP4 (Fig. 5.A), AQP4 density was quantified (Fig. 5.B) and showed significant interaction of diet and genotype effect (8.3% ± 2.2%, p < 0.001) at 12 months. CHOW-fed TgAD rats had higher AQP4 density compared to CHOW-fed nTg rats (by 11.8% ± 1.8%, p < 0.001) and was rescued by the HCHF diet (Fig. 5.B.i), driven by males (10.5% ± 1.7%, p = 0.002) (Fig. 5.B.ii). With prolonged exposure to HCHF diet, there was a female-driven attenuation of hippocampal AQP4 density (significant interactions of diet-genotype-sex: 20.% ± 5%, p < 0.001; genotype-sex: -17% ± 3%, p < 0.001; age-genotype: -14% ± 4%, p < 0.001; and genotype: 12% ± 2%, p < 0.001) and female-driven increase in amyloid angiopathy (significant age and sex interaction: -0.44% ± 0.15%, p = 0.004; and age effect: 0.52% ± 0.09%, p < 0.001) (Fig. 6).
Figure 3: HCHF diet elicts capillary constrictions in non-transgenic animals to the level induced by symptomatic AD pathology. (A) Representative tomato lectin-Texas Red stained images of hippocampal capillaries in CHOW- and HCHF-fed nTg and TgAD 12-month-old rats. (B) Quantification of hippocampal capillary diameter showed significant attenuation in TgAD rats versus nTg on the CHOW diet. The HCHF diet caused a reduction in capillary diameter for nTg, whereas TgAD rats were slightly improved (genotype and diet, p < 0.001). (C) Capillary diameter of CHOW-fed female TgAD rats are significantly lower than female nTg rats. HCHF diet reduced the capillary diameter of nTg rats in both female and male rats and in TgAD male rats compared to CHOW-fed nTg rats. At 12 months, N = 50, where N = 26 (12 nTg, 14 TgAD) were fed with HCHF diet and N = 24 (9 nTg, 15 TgAD) were fed with CHOW.
Figure 4: Prolonged HCHF diet attenuates lectin density in TgAD rats. (A) Representative tomato lectin-Texas Red stained images of hippocampal capillaries in CHOW- and HCHF-fed nTg and TgAD 12-month-old rats. (B.i) 12-month-old TgAD rats have higher lectin density compared to nTg rats. HCHF attenuates the lectin density in TgAD rats and further decreases with prolonged HCHF diet. (B.ii) Both male and female TgAD and nTg rats have comparable lectin density at 15 months. At 12 months, N = 50 where N = 26 (12 nTg, 14 TgAD) were fed with HCHF diet and N = 24 (9 nTg, 15 TgAD) were fed with CHOW. At 15 months, N = 17 (7 nTg, 10 TgAD) were fed with HCHF diet.
Figure 5: HCHF rescues elevated hippocampal AQP4 density in symptomatic AD. (A) Representative stained images of AQP4 in both CHOW- and HCHF-fed 12- and 15-month old (A.i) male and (A.ii) female rats. (B.i) at 12 months of age, there was a significant interaction of genotype and diet (p < 0.001), whereby AQP4 density was elevated in TgAD rats, but rescued with HCHF diet and maintained with extended HCHF diet at 15 months. (B.ii) The attenuation of hippocampal AQP4 density of TgAD rats (by -13.1% ± 1.7%, p < 0.001) at 12 months is driven by males. With prolonged exposure to the HCHF diet, there was a female-driven attenuation of AQP4 density (by -9.30% ± 1.9%, p = 0.028). At 12 months, N = 50, where N = 26 (12 nTg, 14 TgAD) were fed with HCHF diet and N = 24 (9 nTg, 15 TgAD) were fed with CHOW. At 15 months, N = 17 (7 nTg, 10 TgAD) were fed with HCHF diet.
Figure 6: Prolonged HCHF exacerbates cerebral amyloid angiopathy in females. (A) Representative stained (tomato lectin-Texas Red) images of hippocampal CAA in TgAD rats fed the HCHF diet for 3 months (12-month-old rats) and 6 months (15-month-old rats). (B.i) Quantification of hippocampal CAA density showed significant interaction of age and sex (p = 0.004) and a significant effect of age (p < 0.001): the increase in CAA coverage with age was driven by female rats. At 12 months, N = 14 HCHF-fed TgAD rats. At 15 months, N = 10 TgAD HCHF-fed TgAD rats.
Attenuation of glucose uptake with age in CHOW-fed cohorts and its enhancement in HCHF-fed TgAD cohort
From the time course of glucose uptake (shown in Fig. 7.B), we computed the area under the curve (AUC) to quantify the amount of 2DG uptake in each cohort over the 60-minute period following the onset of 2DG injection (Fig. 7.C). In light of the kinetics of glucose uptake (Joo et al. 2022), the 60-min glucose uptake time courses were divided into three 20-min segments (as shown in Supplementary Fig. 3), with the initial glucose uptake phase captured by the 1st segment, the peak uptake phase by the 2nd segment, and wash-out phase by the 3rd segment. CHOW-fed rats showed age-dependent attenuation of glucose uptake during the first segment: where 6-month-old rats have lower AUC compared to 12-month-old rats (0.073 ± 0.023, p = 0.002). The glucose uptake attenuation with age in CHOW rats was driven by male rats (age-genotype-sex: -0.345 ± 0.148, p = 0.019; genotype-sex: 0.382 ± 0.123, p = 0.002) (as shown in Supplementary Fig. 4). HCHF-fed rats showed significant interaction of age and genotype effects on glucose uptake (60-min, p = 0.003; 1st, p = 0.003; 2nd, p = 0.013; 3rd, p < 0.001): the 12-month-old TgAD rats had higher glucose uptake (by 0.307 ± 0.075, p = 0.035; 0.157 ± 0.031, p = 0.006 at 3rd segment) when compared to 12-month-old nTg rats and higher (by 0.536 ± 0.092, p = 0.002; 0.187 ± 0.033, p = 0.003 at 2nd segment; 0.177 ± 0.038, p = 0.015 at 3rd segment) when compared to 6-month-old TgAD. The significant increase in glucose uptake of HCHF-fed rats was driven by females (1st segment| age-sex, p = 0.032; 2nd segment| age-sex, p = 0.008; sex, p = 0.019): in particular, female 12-month-old TgAD rats exhibited significantly higher glucose uptake compared to female 6-month-old TgAD rats (2nd segment: 0.294 ± 0.057, p = 0.025).
At 6 months of age, there was a significant interaction of diet and genotype (60-min, p = 0.036) and a significant effect of genotype (2nd segment, p = 0.002): CHOW-fed TgAD rats showed a trend toward higher glucose uptake compared to CHOW-fed nTg (60-min: 0.078 ± 0.123; 1st : 0.005 ± 0.048; 3rd : 0.079 ± 0.043). HCHF-fed TgAD rats, in turn, showed a trend toward lower glucose uptake when compared to HCHF-fed nTg (60-min: 0.181 ± 0.120; 1st : 0.049 ± 0.047; 3rd : 0.039 ± 0.042). The attenuation of glucose uptake of HCHF-fed cohorts when compared to CHOW-fed cohorts were driven by males in nTg and by females in TgAD (60min | diet-genotype-sex, p = 0.008; diet-sex, p < 0.001; sex, p = 0.017; 1st segment| diet-genotype-sex, p = 0.001, diet-sex, p < 0.001; 2nd segment | diet-genotype-sex, p < 0.001, diet-sex, p < 0.001; sex, p = 0.012). However at 12 months of age, there was a significant interaction of diet and genotype (60-min, p < 0.001), diet (1st segment, p < 0.001; 2nd segment, p = 0.014), and genotype (60-min, p < 0.001; 2nd segment, p = 0.026; 3rd segment, p < 0.001): HCHF-fed TgAD rats showed higher AUC (by 0.483 ± 0.1, p = 0.008; 0.175 ± 0.037, p = 0.009 at 3rd segment) compared to CHOW-fed nTg rats, (by 0.129 ± 0.033, p = 0.044 at 1st segment) compared to CHOW-fed TgAD rats and (by 0.157 ± 0.037, p = 0.021 at 3rd segment) compared to HCHF-fed nTg rats. It is important to note that all the significant differences after pair-wise comparison of the different groups persisted in female rats only (Fig. 7.D)
Figure 7: HCHF attenuates hippocampal glucose uptake in presymptomatic AD, but elevates it in symptomatic AD. (A) Representative time-resolved MTR change map at baseline and changes after 2DG infusion (gray bar) at 20 mins interval showing uptake of 2DG. (B) Dynamic CEST ΔMTR (mean and SEM) time courses in 6-, 12-, and 15 month-old nTg and TgAD cohorts fed with CHOW or HCHF diets. The time origin is the start of the 6-min 2DG infusion, indicated by the gray-shaded rectangle. (C) Area under the curve (AUC) of all scanned rats and (D) AUC of only female rats over 60 minutes after the onset of 2DG infusion for 6-, 12-, and 15-month-old cohorts. There was a significant interaction of diet and genotype in 6- (p = 0.036) and 12-month-old (p < 0.001) cohorts. At 6 months of age, HCHF-fed nTg rats had − 0.122 ± 0.115 lower AUC compared to CHOW-fed nTg; while HCHF-fed TgAD rats had − 0.381 ± 0.127 lower AUC compared to CHOW-fed TgAD rats. However, at 12 months of age, these contrasts were reversed: HCHF-fed nTg rats had 0.176 ± 0.095 higher AUC compared to CHOW-fed nTg; while HCHF-fed TgAD rats had 0.352 ± 0.102 higher AUC compared to CHOW-fed TgAD. Considering the HCHF-fed cohorts, 12-month-old TgAD rats showed highest levels of hippocampal glucose uptake: they exhibited higher AUC compared to 12-month-old nTg rats (by 0.307 ± 0.075, p = 0.035), to 6-month-old nTg rats (by 0.536 ± 0.092, p = 0.0002), and to 15-month-old nTg rats (by 0.392 ± 0.096, p = 0.035). (D) CHOW-fed cohorts at 6 months, female nTg rats have a trend of higher glucose uptake compared male nTg rats while male TgAD rats have a trend of higher glucose uptake compared to female TgAD rats. With exposure of 6-month-old cohorts to HCHF diet, the attenuation of glucose uptake in nTg rats was driven by female rats and TgAd rats by male rats (sex: p = 0.017; diet-sex: p < 0.001; diet-genotype-sex: p = 0.008) when compared to CHOW-fed littermates. The enhancement of glucose uptake in the 12-month-old TgAD cohort compared to 6-month-old TgAD cohort with HCHF diet is largely driven by females (1st segment: age-sex, p = 0.032; 2nd segment: age-sex, p = 0.008 and sex, p = 0.019). Specifically, HCHF-fed 12-month-old TgAD female rats have significantly higher glucose uptake (by 0.294 ± 0.057, p = 0.025 in the 2nd segment) compared to 6-month-old TgAD female rats. In addition, the attenuation of glucose uptake of HCHF-fed 15-month-old TgAD rats compared to 12-month-old TgAD rats was driven by females (2nd segment: age-genotype-sex, p = 0.020) but not statistically significant after pairwise comparison.
Attenuated resting CBF and enhanced functional hyperemia at 6 months; preserved resting CBF and rescued functional hyperemia of TgAD rats at 12 months after HCHF diet
Representative activation maps and CBF time course change in response to bilateral forepaw stimulation of CHOW- and HCHF-fed nTg and TgAD cohorts at 6, 12, and 15 months are shown in Fig. 8.A. CHOW-fed rats showed significant attenuation of resting CBF with age: 12-month-old rats have lower resting CBF (by -31 ± 7 ml/100g/min, p < 0.001) compared to 6-month-old rats. 12-month-old TgAD rats had significantly lower resting CBF (by -57 ± 13 ml/100g/min, p = 0.033) when compared to 6-month-old nTg rats. At 6 months of age, HCHF-fed rats had significantly lower resting CBF (by -34 ± 8 ml/100g/min, p < 0.001) compared to CHOW-fed rats, but resting CBF among cohorts, fed with either diet, was indistinguishable at 12 months of age. Notwithstanding, there was a significant sex effect in resting CBF at 12 months of age (p = 0.010): CHOW-fed rats showed a trend toward lower resting CBF, driven by males; and HCHF-fed female rats had lower resting CBF compared to male rats in both nTg and TgAD cohorts. With prolonged exposure to the HCHF diet (at 15 months), there was a male-driven attenuation (p = 0.016) of resting CBF at 15 months.
HCHF-fed rats showed higher CBF response to forepaw stimulation (by 7.4% ± 1.2%, p < 0.001 at 6 months; 5.5% ± 0.8%, p < 0.001 at 12 months) compared to CHOW-fed rats. Considering CHOW-fed rats, TgAD rats had higher CBF responses (by 4.2% ± 1.3%, p = 0.001) when compared to nTg rats. The area of activation - assessed by the ratio of activated voxel in response to stimulation - showed that at 6 months of age, HCHF-fed rats had larger areas of activation (by 0.17 ± 0.04, p < 0.001) when compared to CHOW-fed rats. In addition, TgAD rats had larger areas of activation (by 0.11 ± 0.052, p = 0.035) compared to nTg rats. However at 12 months, there was a significant interaction of diet and genotype (p = 0.009) where HCHF-fed nTg rats showed a trend toward lower area of activation (by -0.05 ± 0.07) when compared to CHOW-fed nTg while HCHF-fed TgAD rats showed a trend toward higher area of activation (by 0.18 ± 0.06) compared to CHOW-fed TgAD rats (p = 0.53). Comparing HCHF-fed cohorts at 12 months vs. HCHF fed cohorts at 15 months of age, there was a significant interaction of age and genotype (p = 0.001) with respect to the area of activation: 15-month-old rats had larger area of activation (by 0.20 ± 0.04, p < 0.001) when compared to 12-month-old rats, driven by male nTg rats (p = 0.021).
Figure 8: HCHF elevates functional hyperemia in presymptomatic and symptomatic AD; and decreases resting perfusion in presymptomatic AD. (A) Average hippocampal ASL time courses in each cohort normalized by mean resting ASL and representative maps of CBF changes in response to bilateral forepaw stimulation of CHOW- and HCHF-fed nTg and TgAD cohorts at 6, 12, and 15 months of age. (B.i) CHOW-fed 12-month-old rats had lower resting CBF compared to CHOW-fed 6-month-old rats (30.7 ± 6.7 ml/100g/min, p < 0.001); whereas 12-month-old TgAD rats had significantly lower resting CBF compared to 6-month-old nTg rats (57 ± 13 ml/100g/min, p = 0.033). In addition, CHOW-fed TgAD rats had lower resting CBF when compared to CHOW-fed nTg rats (33.58 ± 15.33 ml/100g/min, p = 0.029). At 6 months, HCHF-fed rats had significantly lower resting CBF compared to CHOW-fed rats (35.5 ± 7.7 ml/100g/min, p < 0.001). (B.ii) HCHF-fed rats showed a greater increase in CBF in response to forepaw electric stimulation at both ages (6 months: 7.4% ± 1.2%, p < 0.001; 12 months: 5.5% ± 0.8%, p < 0.001). (B.iii) At 6 months of age, HCHF-fed rats had a larger area of activation when compared to CHOW-fed rats (0.171 ± 0.043, p < 0.001); whereas HCHF-fed TgAD rats had significantly higher area of activation compared to CHOW-fed nTg rats (6 months: 0.281 ± 0.053, p = 0.012). At 12 months of age, there was a significant interaction of diet and genotype (-0.244 ± 0.094, p = 0.009): HCHF-fed nTg rats exhibited a trend toward lower area of activation when compared to CHOW-fed nTg (0.047 ± 0.055). In contrast, HCHF-fed TgAD rats showed a trend of higher area of activation compared to that of CHOW-fed TgAD rats (0.183 ± 0.058). At 15 months of age, there was a significant interaction of age and genotype (0.214 ± 0.0628, p = 0.001): HCHF-fed 15-month-old rats had larger area of activation when compared to HCHF-fed 12-month-old rats (0.197 ± 0.041, p < 0.001). (C) The observed difference in resting perfusion is driven by male rats while the differences in functional hyperemia are driven by female rats.